[TowerTalk] NEC 5.0 ???

[Máximo EA1DDO_HK1H]

Hi Jim,
As far as I know, for Cubical Quads, MiniNEC (Mmana) gets more accurate results than NEC2/4.
I haven't tested myself but it's what I remember to read somewhere else.

Do you know how NEC5 performs on same Cubical Quads?

Thanks

73, Maximo - EA1DDO

<http://foro.ea1ddo.es/>


________________________________
De: TowerTalk <towertalk-bounces at contesting.com> en nombre de Lux, Jim <jim at luxfamily.com>
Enviado: viernes, 12 de marzo de 2021 1:39
Para: towertalk at contesting.com <towertalk at contesting.com>
Asunto: Re: [TowerTalk] NEC 5.0 ???

On 3/11/21 5:03 PM, Artek Manuals wrote:
> Chuck
>
> The person I spoke to at LLL said the  $110 includes an Individual (
> non commercial use) license , a copy of the "Windows Executable" and
> the manual to go with it
>
> You can still get 4.2 if you want it $300 of an individual license,
> and a copy of the "source code". You then I guess have to write you
> own GUI and compile it or shell out $675 to W7EL net out of pocket $975
>
> I ll probably spring for the $110 and hope the "windows executable" is
> useful to us mere mortals
>
> Dave

NEC4.2 comes with windows executables and a variety of graphical
interface programs (NECPlotG, PatplotPC), but as far as I can tell, no
handy "geometry editor".


Of course, 4nec2 works really well with NEC4.2, and is free.   If
they're licensing NEC5 for $110 for individuals, they might well license
NEC4.2 for the same price. Or, they want you to fork out the $300 -
they're really not in the "money making" business, but the hoops to set
a price are numerous and many, and they probably don't want to fool with it.

It does come with complete source code, compilable for a variety of
targets. I've compiled it for both Linux (Ubuntu, Debian, and RH) and
for MacOS (Mojave and Catalina) using both gfortran and the Intel
compilers.  It does assume you have the usual high performance libraries
(LAPACK) which are freely downloadable. Getting it to compile isn't as
simple as "make clean; make" but it's pretty close.


*Individual:*$110
*U.S. Academic:*$400
*Non-U.S. Academic:*$600
*U.S. Small Business:*$1,200
*Other U.S. Entity:*$1,700
*Other Non-U.S. Entity:*$2,400

https://ipo.llnl.gov/technologies/software/nec-v50-numerical-electromagnetic-code



https://ipo.llnl.gov/sites/default/files/2020-07/NEC5%20Validation%20Manual%20092419.pdf

describes the differences -> for *wire antennas and Yagis* NEC4 might
actually be better.

"Many of the results here were chosen to illustrate known limitations of
NEC–4, and NEC5 is seen to be more accurate for most situations. One
case where NEC–4 beats NEC–5 is in convergence for a dipole, since NEC–5
uses linear, triangular basis functions while NEC–4 has a higher order
sinusoidal basis. The mixed-potential code should converge faster with a
higher-order basis, but there are no plans to add that to NEC–5. So
NEC–4 may remain the preferable code for modeling antennas such as Yagis
where accurate resonant frequencies are critical"

If you have small loops (Circumference<wavelength) near structures or
wires connected to surfaces, NEC-4 does have problems. NEC5 or a
specialized version of NEC-4 (NEC-4SL) don't have this problem.  (Jerry
wrote this up in 1987, I don't know if the code is available or not)

_______________________________________________



_______________________________________________
TowerTalk mailing list
TowerTalk at contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk